Slashdot Mirror
Toyotas Suddenly Accelerate; Owners Up In Arms
cyclocommuter writes "Some Toyota owners are up in arms as they suspect that accidents have been caused by some kind of glitch in the electronic computer system used in Toyotas that controls the throttle. Refusing to accept the explanation of Toyota and the federal government (it involves the driver's-side floor mat), hundreds of Toyota owners are in rebellion after a series of accidents caused by what they call 'runaway cars.' Four people have died." The article notes: "The National Highway Traffic Safety Administration has done six separate investigations of such acceleration surges in Toyotas since 2003 and found no defect in Toyota's electronics."
I have to say that the decline in manual transmission driving has really diminished people's driving abilities. It's one thing that the there's an acceleration issue. It's another thing to not consider putting the car in neutral when something like this is encountered.
And again, nothing was ever found to be wrong with the cars. Seems most of the drivers were used to American cars, and the Audi had both brake and accelerator a little to the right of the more typical position. They were pressing the accelerator instead of the brake. Fact is, in almost all commonly available cars, if you stand on the brake and on the accelerator simultaneously, the car will go nowhere. For events to have happened as described, you'd need the simultaneous failure of two unrelated systems, which both healed themselves miraculously after the event. Additionally, same as last time, there are a few unfortunate cases followed by a deluge of similar claims. I wonder why...
My Hyundai nearly always shows 3mph higher than my GPS. My GPS always agrees with the street-side radar things that flash your speed at you exactly. I never actually compared the long distance odometer readings to actual distance traveled to figure out if the speedometer needle is just aimed a few degrees too far to the right, or if the calibration of the axle rotation to distance traveled is just slightly off. I suppose it could be either one, but it's less evil to think that auto manufacturers just want you to drive slower and saver by tricking you into thinking you are going faster than you actual are that it is to think that they are trying to inflate their MPG figures.
Morphing Software
I'd think not. It's called "drive by wire" technology. I bought an '09 Civic and the thing has a sensor attached to the gas pedal instead of the traditional wire directly to a butterfly valve. I read somewhere that if the PCM didn't think things were right, there's a failsafe "limp home" mode that trips the throttle plate to some slightly higher than idle position and disconnects the pedal and any other controls. One of the sales guys who I met in the process of buying this car insisted that the computer controlled throttle makes it more responsive and safer. Throttle by wire IMO, is fucking with the KISS rule.
I saw a TV show where they interviewed people this happened to. One was on a test drive! Another recalled the smell of the brakes burning, as the motor has so much torque at 0rpm that the brakes are useless. One even put the transmission in park, to no avail. None of the stories sounded like a moron who floored it but thought they were braking.
Of course the computer systems weren't designed to diagnose a problem like this, so there is no record of things like speed vs. pedal position as there would be in an airplane. This means that the dealership runs the standard self-check diagnostics and declares that nothing is wrong with the car, so the driver must have been mistaken. But that's like taking your computer to get fixed after some software crashed. Obviously whatever caused it to crash isn't going to be around anymore, yet nobody tells the user that it couldn't have crashed!
Nobody's saying that floormats don't get stuck over the gas pedal every so often (it happened to me once), but there are other cases that floormats and wrong pedals don't explain. And those need to be investigated and fixed.
dom
I have a friend with a car that has a drive by wire throttle, to facilitate the traction control. That car is now having a problem when accelerating. It's only at particular throttle positions, but it acts funny. My car with a good ol' fashion cable between the pedal and throttle body is very very reliable. If the cable goes, I can replace it fairly cheap. It's much more expensive to replace the more exotic parts.
Serious? Seriousness is well above my pay grade.
I guarantee you this is another example of driver error in the same vein as the unintended acceleration that afflicted Audi 5000's years ago. If I'm not mistaken I think the problem in the Audi was that the position of the pedals was slightly off from what people were accustomed to causing them to think they were pressing down on the brake when they actually had the accelerator down to the floor. There have been a few other cars with similar issues.
I'm quite certain the problem with these Toyota's is similar. How in the hell could a car possible start accelerating on its own? And even if the accelerator is drive-by-wire the brakes are not and will likely never be. This means that if the owner got on the brakes hard they'd be able to slow the car. Even if the ECU didn't cut power when braking as some cars do, the engine won't be able to overpower the brakes. About the only possible culprit I see is cruise control, but again, that should be fairly easy to defeat.
The fact is that when some people panic they freeze up and are unable to do anything else. As with the Audi, they press the gas accidentally, the car lunges forward and they panic, pressing down harder on the pedal. It reminds me of what happened to my father years ago. He was teaching my sister's friend to drive. For whatever reason she got on the gas, started barreling towards a car and hit it. She freaked out and froze, her foot firmly planted on the gas. My father actually had to take her leg and lift it off the gas because she was completely unresponsive.
And the problem is that sometimes the issue isn't actually unintended acceleration but some other problem that gives that impression. I know of some cases, for example, where a transmission doesn't engage properly for whatever reason. The driver tries to accelerate but the car doesn't move, so they give it more gas. The transmission eventually does engage and the car lunges forward more aggressively than anticipated. The car may have a real problem, but the driver didn't respond to the issue appropriately.
People nowadays are far too ignorant about they drive. Some people barely know what they're driving, let alone how anything works. As part of driver training basic instruction on the mechanical operation of a car should be mandatory. This would allow drivers to better respond to problems and make them better informed when they deal with mechanics so that they don't get taken advantage of so easily. It's like Toyota's recall over the floor mats. Are drivers so oblivious that they don't notice their floor mats riding up under the pedals. It's not like those things slip under there that easily. Too many people seem to take driving as seriously as they do sitting on the sofa watching television. But they sure do manage to have quite an ego about what they drive.
...It's called "drive by wire" ...
and it's nicknamed 'die-by-wire'.
Never shake hands with a man you meet in a fertility clinic.
Sorry for the anon posting, but I don't want to undo my moderation in this thread ...
I had this same problem with a 1989 Jeep Cherokee in 2000. People died because of similar problems. On the internet and in court, Jeep claimed it was user error. The problem is people don't know enough about their cars to diagnose it, but it turned out to be the Throttle Position Sensor. Which would randomly rev the engine to 4000 RPM's instead of the idle of ~900 when you put it into gear. Yes, absolutely unpredictably. It was not easily duplicated for a mechanic, because when the TPS first started to go bad, it was very infrequent.
The problem is your normal routine is start the car and put it into gear almost immediately. The engine takes more time to rev noticeably past idle speed, at which time you're already moving and lost control of the car. Jeep claimed this was an unreasonable explanation, because they had an engineer sit in the car, depress the brake, put the car into gear, and then rev the engine and be unable to overpower the brakes. I found this only to be true if I were standing on the brakes, something I wasn't in practice to do, from a stop. It also overlooked the problem that once you let the car start moving, getting it stopped again was extremely difficult.
So everyone always assumes there are enough idiots out there for it to be driver error, but it happened to me, so I never trust any of the car manufacturers when this problem creeps up which it does fairly often. Also, it seems like extremely poor engineering on the manufacturer's part to fail to acknowledge this possible avenue of failure. Seems more like they are just covering their asses to avoid culpability.
I was in college at the time, and this is probably one of the more valuable lessons I've ever learned in engineering and something I think would be valuable in software engineering also. Never, ever dismiss complaining customers as morons just because its the simplest explanation especially regarding a safety issue. People actually put up with a lot. More often than not, when people complain and it is difficult to do so, there is merit to the complaint. A proper investigation is required, and a open mind, and wide imagination help determine the failure states.
Throttle by wire IMO, is fucking with the KISS rule.
To propose that somehow mechanical cables are safer because they are simpler is severely flawed.
Have you every seen how easy it is to get a mechanical throttle to get stuck? Also in a wreck the cable the throttle can become pinched, stuck at full throttle not fun. Look at the back of street/strip drag car they have emergency kill switch on the back bumper for this reason, along with wanting to kill the fuel pump.
A mechanical throttle lacks any safety controls. A simple cable is in fact simple, but it is a stupid cable.
Twice I have seen a mechanical fail.
90s Camaro SS , guy had dropped about 10k in the motor only to have the throttle get stuck, redline, broken valves.
My Saturn has become stuck more than once when it is cold, never caused an accident.
Just because you drive car,bought a car, doesn't give you divine insight on how a car works, and somehow your experience brings insight to the conversation.Spend a little time working on/ modifying a car and you will quickly discover; mechanical cables suck ass to put in and are more prone to failure!
"Mechanical and hydraulic systems are much better understood."
Computer aren't exactly a new-fangled technology, and most cars built after the 70's or so have had some form of by-wire throttle control (I'm speaking, of course, of electronically controlled anti-lock brake systems which, while not fully by-wire, include a by-wire system alongside the mechanical one). I always find it amazing how people can talk about computer controlled systems are not being 'not well understood' in this day and age. Not well programmed, perhaps (though the analog there would be poorly fabricated parts, a problem that was dealt with much as poorly coded systems are being dealt with now-a-days), but not well understood? Really?
This isn't a new technology, it's been around for a while in one form or another. Sure it's less mature than conventional hydraulic, but hydraulic are much less mature than banging two sticks together, doesn't make the sticks better than hydraulic. I think my signature sums it up quite nicely, just because drive-by-wire is something relatively new (50yrs of use) and hydraulic something relatively old doesn't make one better or worse than the other, that depends on how they function. If you want to compare the two systems that way be my guest, but just stating that drive-by-wire is worse because it's newer is hogwash.
There are two kinds of fool One says 'This is old therefore good' Another says 'This is new therefore better'- Dean Ing
It is so that they cannot be sued for displaying a lower speed than the real one - for which reason you could get a damn ticket. And that happens everywhere (or at least "almost" everywhere...)
Never had your accelerator cable become stuck then? I think that if you do the numbers you'll find that a fully mechanical system has a far higher chance of an unsafe failure mode (eg butterfly valve stuck open) than the 'fly by wire' systems.
Yep, and we have *such* and exact value for Pi.
Not to mention the radios of wheels are a perfect constant because tire-tread never wears out, and nobody ever uses non-standard wheel-sizes. Also, wheels never skid even for a moment, when turning there isn't a moment where two wheels are actually going the opposite direction....
In short, accurate speed measurement is basically impossible from inside a car, there is a huge margin of error. In the interest of safety, car manufactures design their speedometers to overcompensate, so the margin is always above true speed. Reporting 5% under true speed would be much more dangerous than 5% over.
One can only get a truly accurate speed measurement using an external measuring device.
Unicode killed the ASCII-art *
F-22, F-35, Boeing - flown by professionals
Premium class auto driven by morons
Two different things. Different environments and different safety measures
If the cable goes, I can replace it fairly cheap. It's much more expensive to replace the more exotic parts.
Also, if the cable gets wet, you have a wet cable. If the leads to the potentiometer have cracked insulation and get shorted out by dirty water or contact with metal, you have a hard open throttle. I know which I'd rather have.
This is why drive-by-wire stuff is scary. Power assist is fine, but if your hydraulic power steering loses pressure you can still steer the car (even if the steering is much heavier). Vacuum-assisted braking has enough storage to last 2-3 brake pedal pushes, and after that the brakes still function even if they too are much heavier. Antilock braking is scary enough even though it's comparatively failsafe. Handing over complete control of any further automotive function to a computer is goddamn terrifying.
Rampant carbon sequestration destroyed the Dinosaurs' tropical paradise. I'm here to help repair the damage.
People can also build up rational fears of what they do understand. There's a lot of that on slashdot, too. Overcomplicated overengineering is something much of slashdot has experience with.
That's all I'm saying... Why I am supposed to care about some malady that befalls you in your foreign car, when my city is destroyed by Big 3 downsizing. If our unemployment is our fault for building shitty cars, then, certainly your accident is your fault for buying one too. Don't come crying to me about your Toyota problems...If you get killed in that thing, its not my problem. You abandoned my country decades ago.
This is my sig.
which is good for averaging but NOT good for spot measurements
Spot measurements? For speed? GPS does the same as any speed measurement - calculates the time it takes to cover a prescribed distance, if you know of another way I'd be interested ;)
If I were programming/designing the drive by wire system you describe (don't worry, I'm not), I'm pretty sure that I would read a shorted potentiometer as closed and only use about 1/3 of the range of the thing (so it would be apparent if it failed in the other direction).
Going a little further, I would only use about 1/3 of the range of it, in the middle, so a short would be an obvious failure condition (as would the highest resistance of the potentiometer).
Nerd rage is the funniest rage.
Pumping the brakes always causes brake failure in cars with power assist braking. It always has. That's why no one teaches you to do it anymore.
For those that don't know, power assist brakes generally get their assist from engine vacuum. There is more vacuum when the throttle is closed (engine idling) than there is when the throttle is open (full acceleration). Normally there's enough stored energy to give one or two assisted full brake applications. After that you run out of assist and are completely manual. Even standing on the brakes might not be enough to overpower the engine, particularly if the driver is a petite woman. Try pumping the brakes with the engine off to see the loss off assist.
The recommendation to not pump the brakes comes from inclimate weather. Cars with ABS it's the wrong thing to do. Step on the brake and let the ABS do the pulsing for you. Unfortunately many drivers don't realize that the pulsating pedal is normal in ABS cars and let up, or start pumping. In cars without ABS "threshold" braking is preferred, but it's a different technique that randomly pumping the brakes, but that's what people take away.
As far as using the hand brake in this situation... good luck. In my car I can fully apply the hand brake, and the rear wheels will lock up, but I can continue to accelerate down the road.
The truth is a shocking number of people don't know how to fully control their car. Given that it is capable of killing themselves or others if it's out of control, that's quite frightening.
I sure hope you've contacted your dealer, the government, and a lawyer.
Give me Classic Slashdot or give me death!